Social media and the transformation of brand
communication
William Ryan
wryan@ithaca.edu
Adam Peruta
aperuta@ithaca.edu
Suzanne Chouman
schouma1@ithaca.edu
Ithaca College
953 Danby Road
Ithaca, NY 14850
Ithaca College
953 Danby Road
Ithaca, NY 14850
Ithaca College
953 Danby Road
Ithaca, NY 14850
ABSTRACT
In this paper, we examine the ways brands communicate via
social media. Using a content analysis, we analyzed 44 of the
world’s largest brands according to Interbrand to discover the way
they communicate through Facebook wall posts and Twitter. We
observed some brands are doing a great job creating an interesting
environment by posting content specifically to the brand
community, but many brands could benefit from creating a more
participative and engaged environment in both Facebook and
Twitter, as well as being more responsive to their fans. These fans
expect they can develop personal relationships with brands on
social media. Maintaining an engaging and responsive
environment will become increasingly important for those who
want to be at the forefront of the social media revolution and reap
benefits such as brand loyalty and trust of healthy and engaged
brand communities.
Categories and Subject Descriptors
K.6.m [Management of Computing and Information Systems]:
Miscellaneous
General Terms
Management, Design
Keywords
Social media, brand communication, audience engagement, brand
communities
1. INTRODUCTION
Social media can be the ultimate game changer for brands,
whether they wish to participate or not. Not only do brands need
to restructure marketing messages to fit a number of emergent
media, but these media fundamentally require a new way to plan
interactions with, monitor, and respond to their consumer bases.
These new media platforms thrust new rules upon brands—the
willing and unwilling as well as the aware and the unaware. This
new branding environment requires action from brands and their
corresponding organizations as consumer expectations are rapidly
transforming and new opportunities for connecting with those
consumers are emerging.
In prior research, we studied how brands represented themselves
on Facebook, comparing those representations to how they
represented themselves on their websites [10]. The focus of that
Permission to make digital or hard copies of all or part of this work for
personal or classroom use is granted without fee provided that copies are
not made or distributed for profit or commercial advantage and that
copies bear this notice and the full citation on the first page. To copy
otherwise, or republish, to post on servers or to redistribute to lists,
requires prior specific permission and/or a fee.
Academic MindTrek 2013, October 1-4, 2013, Tampere, FINLAND.
Copyright 2013 ACM 978-1-4503-1992-8/13/10…$10.00.
study was primarily on the visual design of each. We found
brands were not using the same visual design resources for social
media they were for their websites. We also found some evidence
that slightly less prominent brands are putting more effort into
cultivating their social media presences than slightly more
prominent brands. We compared these top brands to a selection of
up-and-coming institutions of higher education and found these
big brands were using significantly less visual design resources
than even these institutions of higher education, even though
consumers are flocking to social media [11]. But, these studies
both focused solely on visual aspects of the brand. Social media
brand management though requires more than just a “set it and
forget it” mentality. Social media is first and foremost an
interactive, communicative medium consumers use to
communicate with friends, family, co-workers and now brands
they care about.
We wish to explore the interactive nature of social media and how
big brands are responding to consumers. We want to understand
the consequences of maintaining a brand presence. Maintaining
this interaction involves aspects of content generation and
management, responding to participants, monitoring activity,
cultivating a community of followers of the brand, as well as
numerous other responsibilities.
For this study of how brands engage with consumers via social
media, we identified five factors to analyze: content generation,
engagement with the audience, responsiveness, integration, and
policy generation. We derived these factors from the literature
below and aspects of communication we wished to learn more
about. To address these issues, we have established 5 research
questions:
1.
2.
3.
4.
5.
What is the nature of content being posted on these
sites?
What is the relationship between being responsive
to consumers, engaging with the brand’s audience,
and posting content to keep consumers engaged?
What is the relationship between having public
policy for using social media, integrating social
media together, and interacting with consumers in
social media?
What separates active, passive, and inactive social
media interaction?
Are there any differences in how brands approach
Twitter and Facebook?
In this paper, we will explore these relationships and what they
mean for brands trying to maintain a Facebook or Twitter
presence. We start by describing the relationship between brand
management, social media, and brand communities already
discussed in the literature. We then describe the content analysis
to study big brand’s Facebook and Twitter pages. We reveal the
findings we have focusing on the 5 questions above. And, finally,
we discuss what these findings mean for organizations using
social media.
2. LITERATURE REVIEW
We looked into research relevant to this study of how brands are
facing the drastic changes involved in marketing practices due to
the emergence and proliferation of social media. [2] describes a
major transformation happening in digital marketing in general.
Consumers want personal relationships with brands even though
they have a very diverse relationship with many different brands.
The author contends the marketing process should be considered a
decision journey by the consumer. Marketers have an ability to
influence this journey, but only by taking the consumer’s brand
experience into account.
[9] mention the dilemma facing marketers as one between the
desire to control their audience and the audience’s desire to have
an authentic, individual experience with the brand. They mention
the user’s experience as the key to unlocking the potential of
social media—one of the most important is having a community
in which to express thoughts. Although control is given up, [5]
describe this idea that brands have total control over the messages
they seek to diffuse as a fallacy. They argue the internet, due to
social media, is no longer a place of just information, but more
importantly of influence. Giving up this control is a means of
interacting with the audience and encouraging participation and
engagement. They propose using an ecosystem approach of both
social and traditional media to actively engage consumers in
dialog. This ecosystem matches the integration factor we have
sought to understand.
[8] demonstrates how maintaining an effective social media
presence can positively impact consumer brand loyalty. They
describe brand trust as being the mediator for such loyalty. The
findings also confirmed the notion that consumers engaged with
other consumers even within the domain of brand pages. This
means brands need to maintain authentic, open lines of
communication to take advantage of a boost in brand loyalty.
Both [8] and [7] describe effective social media communication as
the ‘glue’ necessary for successful brand management. [7] also
suggest while duration may be appropriate in other media for
understanding audience engagement, frequency is a more valuable
metric in social media (e.g., it is not about how long they stay on
the Facebook page, but how many times they come back). This
frequency also connects with responsiveness, not just of the
consumers, but also of the brand itself.
A concept discussed frequently in the literature relating to brands
and social media are brand communities. [12] explain brand
communities as a form of community found around a brand based
on a triadic relationship with the consumer, other consumers, and
the brand. From the perspective of uses and gratification theory,
the authors explain how users use such brand communities for
functional (e.g., sharing information and opinions), as well as
social (e.g., connecting with the brand and others) and
entertainment (e.g., having memorable experiences) purposes. [8]
argue brand communities play an important role in expressing a
person’s identity; consumers join brand communities to say
something about themselves. On social media sites, they argue,
content is created through participation in the brand communities.
[6] argue brand communities evolve not as a separate
communication channel, but as a medium for devoted consumers
to develop personalized relationships with the brand. Creating a
survey, contacting those engaged within such communities, the
authors found the amount of information exchange between
community members and the brand as well as the monetary and
psychological rewards for participating in the community had the
biggest impact on brand loyalty. Finally, [4] show how brand
communities offer a financial benefit for brands with healthy
communities. They show how engagement in social media brand
communities increases purchasing behavior within consumers.
They also show user-generated content (as opposed to marketergenerated content) has a stronger influence on that behavior.
Finally, the communicative nature of the media has had some
focus from a brand communication effectiveness standpoint in the
literature. [1] studies the quality and nature of brand post
effectiveness. They identify that the factors influencing likes of a
page are not the same influencing the number of comments for a
post. Their findings are that the vividness of posts and inclusion of
some interactivity on a wall post, as well as positive comments
from others positively affects a page’s likes, while high
interactivity and original brand content have the highest
connection with the number of comments for a post (as well as
both positive and negative comments and discussion from others).
[13] extend further this analysis of the communication function in
social media as differentiated by platform. YouTube typically
features brand messages the most about the brand itself, while
Twitter was the least (e.g., the brand is background to the
interaction). They assert Twitter messages revolve around news
and information sharing, Facebook around social connectedness,
and YouTube around self-presentation. This research again shows
brand communication must occur across different social media
channels yet remain cohesive and integrated.
Clear themes emerge from this literature for successful social
media brand communication strategies. Brands have the ability to
influence the environment in which communication about the
brand takes place, but brand loyalty and other rewards for the
brand can only be found in supporting the community to form
itself organically and authentically. This research shows social
media is at its most effective when brands allow loyal consumers
to be the stewards of the brand in these communities. The brand’s
role is to structure the environment in which consumers
communicate, provide content engaging people with the brand’s
themes and values, and be ready always to respond to the
community or individuals when appropriate. The transformation is
not just a shift from traditional marketing techniques, but a
fundamental change in the philosophy of marketing focused on
creating lasting relationship with loyal consumers.
3. METHDOLOGY
For our study, we performed a content analysis to analyze how
brands are handling communication through social media. We
focused specifically on communication through Facebook and
Twitter. Forty-four brands in total were randomly sampled from
Interbrand’s annual list of 100 Best Global Brands. Our selection
was made from the 2011 listing. The brands listed in Interbrand’s
report are ranked according to the ongoing investment and
management of the brand as a business asset. The Interbrand
methodology takes into account all ways in which a brand touches
and benefits its organization—from attracting and retaining talent
to delivering on customer expectations. The three aspects
contributing to the Interbrand assessment are the financial
performance of the branded products or services, the role of brand
in the purchase decision process, and the strength of the brand
overall. Table 1 lists the selected brands with the associated sector
in descending order alphabetically.
Table 1. Brands used in the content analysis. Asterisks mean
the brand was analyzed by two different coders.
Brand
Interbrand Ranking
Sector
Zara
Accenture
45
Business Services
Adidas
60
Sporting Goods
Amazon.com
26
Internet Services
Avon*
65
FMCG
Barclays
79
Financial Services
Data was collected January 2013 from brands’ Facebook and
Twitter pages posted during two separate four-day spans of
Wednesday, November 21, 2012 to Saturday, November 24, 2012
and Sunday, December 2, 2012 to Wednesday, December 5, 2012.
This provided two separate sample points, preventing any single
event (e.g., Thanksgiving holiday) from providing misleading
patterns in the data.
Blackberry
56
Electronics
BMW
15
Automotive
Budweiser *
29
Alcohol
Coca-Cola
1
Beverages
Colgate
51
FMCG
Corona
86
Alcohol
Dell*
43
Technology
Disney*
9
Media
Ford
50
Automotive
Gap
84
Apparel
GE
5
Diversified
Google
4
Technology
Harley Davidson
100
Automotive
Honda
19
Automotive
HP
10
Technology
Hyundai
61
Automotive
IBM
2
Business Services
Ikea
31
Intel
44
Apparel
Our data collection form included some data about the pages
themselves such as the number of likes, followers, and following.
We also collected a great deal of data about the wall posts and
tweets, including the number of each, number of each directed at
the brand by other users within the past 24 hours, and the number
of responses. We also looked for the presence of links to other
social media pages and public social media policies.
Table 2. Overall page data collected
Facebook (count)
Twitter (count)
Link to.. (y/n)
Likes
Followers
Facebook Acct. on
homepage
Talking Abouts
Following
Twitter Acct.
homepage
on
Posts (first dates)
Tweets (first dates)
Twitter Acct.
Facebook
on
Posts
dates)
Tweets
dates)
Facebook Acct. on
Twitter
(second
(second
Posts in last 24
hours
Tweets in last 24
hours
Home Furnishings
Responses in last
24 hours
Responses in last 24
hours
7
Electronics
Social Media Policy… (y/n)
Jack Daniels
78
Alcohol
Present on Website
McDonalds
6
Restaurants
Mercedes Benz
12
Automotive
Morgan Stanley
54
Financial Services
Nescafe
30
Beverages
Nestle
55
FMCG
Nissan
90
Automotive
Panasonic
69
Electronics
Pepsi
22
Beverages
Porsche
72
Automotive
Samsung
17
Technology
SAP
24
Business Services
Smirnoff
89
Alcohol
Sprite
63
Beverages
Starbucks*
96
Restaurants
Tiffany & Co.
73
Luxury
Toyota
11
Automotive
UPS*
27
Transportation
Visa
75
Financial Services
Present
Facebook
on
Facebook
app
Twitter
Present on Twitter
Policy Sections (across all policies) (y/n)
Purpose of social
media usage
Copyright, privacy,
confidentiality
Code of conduct by
online community
Code of conduct by
brand
representatives
Treatment
of
Offensive Posts
The second part of the form was used to perform a content
analysis on the social media content. Our sample consisted of
every tenth wall post and tweet within the date ranges. This
provided a reasonable amount of data collected, while not
overwhelming the researchers as there were an average of 7 wall
posts and 41 tweets for each brand sampled. Each Facebook wall
post in the sample was categorized; media type was recorded; and
the number of shares, wall post character length, and comments
were counted. Each tweet in the sample was categorized and the
time delay was noted if the tweet was a reply or retweet to another
user’s tweet. In total, we analyzed 72 Facebook wall posts and
235 tweets. Table 3 shows the categories and media types for each
Facebook wall post and tweet.
Table 3. Categories and media types for Facebook wall posts
and tweets.
Category
Description
Event
The post/tweet is promoting some timebased activity (online or offline).
Contest
The post/tweet encourages participation
from online community by competing
with each other (whether or not there is a
reward for wining).
Special Promotion
The post/tweet promotes a special offer.
Product Promotion
A product or service is being advertised.
Brand Related
The post/tweet if it makes reference to the
brand itself in some way (visual design,
organization, etc.)
@ Reply
Tweet is a response to another twitter
account or makes reference to another
account.
Retweet
Tweet is prefaced with a RT or marked as
“Retweet” from another Twitter user.
Other
The post/tweet doesn’t fit any other
category.
Media Type
(Facebook only)
Text Only
The post contains only text (not even a
link).
Link
The post contains a link (may include an
image, because it includes a summary of
the link).
Poll
The post contains a question with options
the users can select. May contain a link as
well.
Image
The post contains an image specifically
(and may also contain a caption). May
contain a link as well.
Video
The post contains a video (and may also
contain a caption). May contain a link as
well.
In addition, each tenth user comment on the Facebook posts
included in the sample above was analyzed. Additional sampling
was required since many posts had hundreds or thousands of
comments, resulting in a combinatorial explosion of content to be
analyzed. The comments were categorized as positive, negative,
neutral, positive about a competitive brand, and negative about a
competitive brand. Furthermore, we recorded each time the brand
responded on a wall post through comments, recording the
character length of the comment as well.
The brand Facebook and Twitter pages were located by doing a
simple search using the sites’ respective internal search engines
and typing in the full name of the brand (e.g. “Harley Davidson”),
unless the links were found directly on the homepage. If the pages
could not be located using this method, we visited the brand Web
site by either doing a search using Google or typing the brand
name as a URL directly into a web browser. We were able to
locate all 44 of the sampled brands’ Facebook and Twitter pages
using these methods.
The data collection form was pretested, revised, and used to
collect data from each of the 44 brands. Six brands overlapped so
we could compare reliability data. We employed 4 coders who
were each given an instruction and data collection sheets to
capture data from the study.
4. FINDINGS
4.1 Reliability Data
The intercoder reliability statistic was calculated during both a
pilot study of the coding sheet as well as in the main study using
overlapping brands to calculate the data. The data was broken
down into nominal and ratio components. Nominal data included
all yes and no questions about the presence of social media
policies and links created between the brand’s website, their
Facebook account, and their Twitter account. Ratio data included
the Facebook likes, shares, number of following/followers,
post/tweet counts, and responses to posts/tweets in the last 24
hours. We also included nominal and ratio data from our
Facebook post coding, where Facebook posts could be matched
up. We were unable to include tweet coding because we were
unable to determine easily matching codings between coders.
These data were fed into the ReCal reliability calculator utility
available at http://dfreelon.org/utils/recalfront/ [3]. For the final
study, the nominal data showed a 78.4% agreement and a
Krippendorff’s alpha value of 0.627. The ratio data showed a
Krippendorff’s alpha value of 0.855. All Krippendorf alpha values
showed that the coding sheet used resulted in acceptable limits for
intercoder reliability.
4.2 Descriptive Statistics
To start, we collected the number of likes on Facebook as well as
the number followed by the brand and the number following the
brand on Twitter.
Table 4. Facebook/ Twitter brand data
Mean
Std. Dev.
8718992.58
13031864.52
Following
8614.18
19501.24
Followers
47377.26
1033477.26
Likes
4.3 Research Questions
RQ1. What is the nature of content being posted on
these sites?
Overall, we recorded the average number of Facebook posts
within a week span, average number of characters in those posts,
average number of comments for each post, and the average
number of tweets. Data is displayed in table 5.
Table 5. Facebook/Twitter descriptive data
Mean
Std. Dev.
6.65
7.79
Avg. Post Char. Count
116.15
93.68
Avg. Comments Per Post
219.12
587.89
41.19
68.53
Avg. Sampled Posts
Avg. Sampled Tweets
In table 6, we show the categories of wall posts and tweets. We
further categorized contest, @ replies, and retweets as being
engaged with the audience, whereas special promotions, product
promotions, and brand-related were categorized as being engaged
with the brand itself. Event and other were not categorized as they
are context-dependent. The percentage of Facebook posts engaged
with audience was 10.25%, while the percentage engaged with the
brand was 89.74%. In Twitter, the percentage of tweets engaged
with audience was 59.11%, while the percentage engaged with the
brand was 40.88%.
Table 6. Post/Tweet category frequency
RQ2. What is the relationship between being
responsive to consumers, engaging with the brand’s
audience, and posting content to keep consumers
engaged?
This question touches on issues of content generation,
engagement with the audience, and responsiveness. We have
described the data of content generation and engagement above.
Data concerning responsiveness can be found in table 8. This
included the average number of sampled responses by the brand to
comments on their Facebook posts, the character length of those
responses, the delay between replying to or retweeting a post from
when it was first tweeted, and the average number of Facebook
posts and tweets directed at the brand and the number of replies
by the brand in a 24-hour period.
Facebook
Twitter
Event
2
1
Contest
8
25
Special Promotion
16
26
Avg. Sampled Responses to
Comments
Product Promotion
24
10
Avg. Char. Count
Sampled Responses
Brand-Related
30
47
@ Reply
0
74
Retweet
0
21
11
31
Table 8. Facebook descriptive data
Avg. Facebook Replies
Other
Finally, we looked at the type of media included with Facebook
posts, whether it was text-only or included a link, poll, image, and
video. (Data found in table 7.) While any of the types may have
encouraged interaction, only 3.26% of the posts used text and
polls designed to encourage engagement.
Table 7. Facebook post media type frequency
Frequency
Percent
2
2.17%
Link
15
16.30%
Poll
3
3.26%
Image
67
72.82%
Video
5
5.43%
Text-only
What should be clear from these data is the relatively small
variance from the amount and type of content posted. There was
usually one Facebook posts per day, six tweets per day, posts
were about the brand itself, and usually including an image. There
was a little bit more variance in the Facebook post text length and
number of comments as well as the type of tweet made.
Furthermore, much of the content in both Facebook and Twitter is
meant to engage consumers, but not to encourage participation,
although Twitter has a much higher level of encouragement.
0.12
0.40
109.44
168.98
23.79
67.25
2.02
5.64
5.89
15.20
45.47
159.19
5.66
21.95
for
Avg. Tweets @ Brand (24
hours period)
Avg. Twitter Replies
Std. Dev.
for
Avg. Posts @ Brand (24
hours period)
Avg. Reply Delay
Sampled Tweets
Mean
We defined content generation as the total number of posts in
Facebook and tweets in Twitter, the average Facebook post
length, and the percentage of Facebook images, videos, and links
posted. We left out text and polls because the nature of the content
comes from different sources—the community and the brand’s
participation in that community. We defined engagingness
according to the number of posts meant to engage and promote
participation as defined in the last section, the number of text or
poll posts generated (e.g., 100% - the links, images, and videos %
above)—with their strong connection with engaging the audience,
the number of accounts the brand has followed on Twitter, and
whether the Facebook pages allowed content posted by audience
members. Finally, we defined responsiveness according to the
number of sampled responses to Facebook comments and the
length of those responses, the time it took for the brand to reply to
audience members’ tweets, and the 24-hour response rate to posts
and tweets directed at the brand.
Using Pearson’s r to correlate these factors in the data (and chisquared analyzing whether brands allowed or disallowed audience
wall posts on their Facebook page), we found only a little support
for these relationships in our data. However, we did find a
moderately, positive relationship between the percentage of
engaged posts and the percentage of engaged tweets—r(37)=0.37,
p<0.05. We also found a rather strong, positive relationship
between the percentage of engaged tweets and the number of
Twitter accounts a brand follows—r(42)=0.50, p=0.001. In
addition for responsiveness, we found a very strong, positive
relationship between the number of responses to Facebook posts
and the length of the response, r(15)=0.66, p< 0.01.
Avg.
Length
Using linear regression analysis, we wanted to see how these five
metrics, which have shown correlation among themselves, related
to other factors we looked at. We found a nearly significant
relationship (p=0.051) for the average reply delay to tweets,
r2=0.92, and a significant relationship (p<0.05) for the total
number of tweets. Regression coefficients are recorded in Table 9,
all values significant at p<0.05 level.
Engaging
Percent
Table 9. Regression analysis of factors with metrics from the
study.
B
Std. Error B
β
Average Twitter Response Length
0.92
0.54
-7.0E-4
1.46
-.12
.011
.003
1.000
Constant
38.23
26.39
Following
-0.004
93.37
-1.08
171.67
93.366
.75
Constant
Following
Avg.
Facebook
Response Length
Total Tweets
Engaging
Percent
Response
0.002
0.001
.41
1.35
0.44
.72
Tweet
Avg. Facebook Response to Comments Length
Constant
Avg. Responses
-6.17
59.03
207.85
8.80
This means certain engagement and responsiveness metrics
correlate to one another, while others in general do not. There are
also some certain relationships between factors, particularly
between those related to engagement and responsiveness, but also
in relation to the amount of content put on Twitter.
RQ3. What is the relationship between having public
policy for using social media, integrating social
media together, and interacting with consumers in
social media?
Below, we have summarized the data having to do with public
policies on social media usage and integrating social media
together with the brand’s website. Table 11 contains frequency
data having to do with integrating websites, Facebook, and
Twitter pages. Table 12 describes the presence of social media
policies available through their various pages and what can be
found in those policies.
Table 11. Frequency data from website, Facebook, and Twitter
Tweet
We also looked at these five metrics through regression analysis
themselves. Following (r2=0.68, p<0.01), engaging tweet percent
(r2=0.80, p<0.001), average Facebook response to comments
(r2=0.77, p=0.001), and the response length (r2=0.51, p<0.05)
were all significant. Regression coefficients for these four
relationships are in Table 10.
Table 10. Regression analysis of factors between themselves.
B
Std. Error B
on integrating social media
Link from
Facebook
Website
Link from
Twitter
Website
Link from
Twitter
Facebook
Link from
Facebook
Engaging
Percent
7205.67
53269.65
19941.31
Not present
29
19
27
23
19
29
4
46
4
45
to
to
β
Following
2289.33
Present
to
Facebook Twitter App
Constant
Tweet
.80
Engaging Tweet Percent
Twitter
to
Using Chi-squared tests, we attempted to see if there were any
correlations in this integration data. The only one was a
connection between linking to Facebook and Twitter from the
website—X2(1)=40.43, p<0.001.
Table 12. Frequency data from website, Facebook, and Twitter
on social media policy
Constant
Following
Avg. Responses
-0.096
0.080
7.38E-6
0.000
.49
0.34
0.11
.63
Present
Not present
SM Policy on Website
27
23
SM Policy on Facebook
27
21
SM Policy on Twitter
0
50
Policy explains purpose
24
19
Avg. Facebook Response to Comments
Constant
.88
0.32
0.14
Policy
explains
copyright,
privacy, confidentiality
28
15
Policy explains code of conduct
for community
30
13
Policy explains code of conduct
for brand representatives
11
32
Policy
explains
consequences of violation
27
16
for
Again, using Chi-squared, we looked for correlation between
where policies were located and responsiveness, engagement, or
content generation, finding only relationships with the average
number of responses to Facebook comments and their length. This
data is found in Table 13.
Table 13. Chi-squared data for location of privacy policy
Policy on Website
Policy on Facebook
Avg. Responses
X2(2)=7.37, p<0.05
X2(2)=3.68, p<0.05
Avg. Response
Length
X2(7)=17.00,
p<0.05
X2(7)=13.17,
p=0.068
We also looked for correlations between where the various policy
content was found. Copyright, privacy, and confidentiality was
found usually on the website—X2(1)=18.05, p<0.001. Meanwhile,
purpose of social media (X2(1)=18.29, p<0.001), code of conduct
for the community (X2(1)=16.59, p<0.001), and consequences for
rule violation (X2(1)=9.74, p<0.01) were usually found on the
Facebook page.
Finally, we found only a handful of relationships between any
factors above and policy or integration issues. First, we found a
relationship between the total number of Facebook posts and
copyright statements (X2(13)=23.77, p<0.05) as well as the
average reply delay in Twitter and copyright statements
(X2(9)=12.77, p=0.103) as well as purpose statements
(X2(9)=14.98, p=0.091). Finally, there was a relationship found
between the average number of responses to Facebook post
comments (X2(2)=6.04, p<0.05) and the length (X2(6)=13.33,
p<0.05) of those responses with copyright statement presence.
There was also a nearly significant relationship between the
number of responses and the presence of a code of conduct for
brand representatives (X2(2)=5.88, p=0.053).
While integration was of less importance, social media policy did
demonstrate a relationship between certain policies and where that
policy was found. Though, having a well-defined policy showed
little relationship with a brand’s actual use of social media except
that most policy information having to do with social media was
usually tied to privacy policies and EULAs.
RQ4. What separates active, passive, and inactive
social media interaction?
We grouped brands according to the metrics we defined above for
content generation for, engaging with, and responsiveness to
audience members. Each brand was categorized by metric
according to the percentile of each brand for each variable. The
percentiles used were the tenth—below which were inactive
brands, fiftieth for active brands, ninetieth for high activity
brands, and one hundredth for the most highly active brands. The
metrics were then added together and divided by the number of
metrics included (e.g., 4 for content generation, 5 for
engagingness, and 5 for responsiveness).
When doing this, we found the following average scores out of a
possible 4 points: for content generation, M=2.57, S=0.88; for
engagingness, M=2.06, S=0.61; and for responsiveness, M=1.10,
S=0.89.
We segmented the groups according to those that were inactive,
those that were passive and focused only on posting content, and
those that interacted with their audience. To define inactive, we
looked for any brands having a score of less than 2.5 for all of
content generation, engagingness, and responsiveness. We defined
passive as having at least 2.5 for content generation and less than
2.5 for everything else. We defined interactive as having at least a
2.5 for engagingness. We found 13 inactive brands, 23 passive
brands, and 14 interactive brands.
The interactive brands were composed of 1 with only
engagingness above 2.5, 7 with content generation and
engagingness above 2.5 (e.g., not actively monitoring their
audience’s usage), and 6 with all three above 2.5.
This shows many brands have an opportunity to expand their
approach to using social media, particularly from an interactive
and participation encouragement standpoint.
RQ5. Are there any differences in how brands
approach Twitter and Facebook?
We used 5 paired sample t-tests to compare data between the two
social media accounts, including the percent of engaged
posts/tweets, the total number of posts/tweets, the 24-hour
response rate to audience’s posts/tweets, the 24-hour number of
responses to audience’s posts/tweets, and finally the number of
likes/followers. We treated these accounts as paired because of
their shared connection through one brand using different media.
We found significant differences for the percentage of engaged
posts and tweets, t(38)=-3.38, p<0.01; number of posts and
tweets, t(45)=-3.48, p=0.01; and number of likes and followers,
t(47)=4.55, p<0.001. The 24-hour responses and response rate is
interesting because while there is not a significant difference
between the two, there is a very strong correlation between the
number of 24-hour Facebook responses and 24-hour Twitter
responses, r(41)=0.69, p<0.001, even though there is not a
correlation in the response rates.
These significant differences are largely due to the structure of the
media. In this case, the medium is the message. The way brands
structure the message follows along with the type of medium they
are using. Given that audiences’ of these brands have certain
expectations of the brand communication in these media, it is well
the brands follow these differences.
5. DISCUSSION
In our prior paper [10], we reported brands were, in general,
actively using their Facebook pages, even though they were not
leveraging their brand identity elements as fully in social media as
on their webpages. We have noticed expanding use of brand
identity elements on such pages by the very same group of brands
since that paper was published.
In this paper, we have established to understand further about the
approach these brands take when connecting with their audience.
We found while brands are utilizing their Facebook and Twitter
pages to post content related to their brand, which is an
appropriate and even expected activity for social media brand
pages, a great deal of room remains to expand on the areas of
audience engagement and responding to consumer interactions
with the brand. Much prior research demonstrates the importance
of using social media not just as a venue for broadcasting a
brand’s message, but for constructing carefully crafted statements
to keep the brand’s audience engaged and help them feel like they
are being listened to.
Social media has been described as being central in terms of its
importance to a brand communication strategy [5; 9] and further is
an important touchpoint to consumers. Most importantly, social
media is an environment in which to engage with brand
communities. Our study has demonstrated the importance of
maintaining engagement with this community and staying
responsive to it in addition to posting original content. We
described how, by and large, engagingness and responsiveness go
hand-in-hand. Not only does the brand need to invite consumers
to participate, they need to respond when the consumer engages
with the brand itself.
We demonstrated how rich media may influence the experience of
consumers’ use of a social media page, but it is not sufficient to
maintain the experience. We found similar results to [13] where
there were significant differences in the way brands engage with
participants in Facebook when compared to Twitter.
We also demonstrated the relative unimportance of integrating
social media accounts together and posting public social media
policy toward the way the brand communicates with its
consumers in relation to the goal of brand communication on
social media. We identified how certain types of information were
more prevalent on Facebook social media policies (e.g., the
purpose of a brand’s use of social media, the code of conduct for
the community, and consequences for inappropriate content) and
others more prevalent on website social media policies (e.g.,
copyright, privacy, and confidentiality information). On the
whole, policies showed weak relationships to the behavior of
brands in content generation, engagingness, or responsiveness.
Finally, it is clear certain brands are separating themselves from
others because of their willingness to engage with and respond to
consumers. This requires more resources than those who just post
content, leaving the interaction mostly to other engaged users. As
[2] points out, we have to consider what sort of experience our
audience is having and if it is the right experience for the message
brands are sending. But, as research has shown, this sort of
engagement can have a greater impact on brand loyalty, trust, and
decision-making.
Our research has demonstrated the importance for brand’s
managing social media accounts to stay interactive with their
audience of devoted consumers. Through the use of interactive
means such as asking questions, having contests, promoting user
content through their official means, replying directly to their
posts/tweets in an authentic and timely manner, and simply
finding and following those who are or may be interested in the
brand, brands can show the consumer why each consumer is
important to the brand and help promote a strong, trusting, and
loyal relationship. Finally, brands must understand the medium in
which they wish to communicate, as each medium has its own set
of built-in expectations about how authentic communication
occurs.
6. CONCLUSION
In this paper, we reported on a content analysis performed to
understand how brands are interacting with their consumers. This
analysis focused on both Facebook and Twitter pages for 44
Interbrand-ranked brands. This study demonstrates the importance
of not just posting interesting content to a web page, but creating a
participative and engaging experience, where consumers can feel
comfortable communicating with the brand and other consumers
in an honest, authentic way. We described how some brands are
doing this, while many are still lagging behind. Brands
approaching the transformation of the traditional marketing
function need to look for new ways to communicate with their
brand communities. Social media is changing the way the game is
played.
7. ACKNOWLEDGMENTS
We would like to thank Skyler Mauro and Greg Acquavella for
their assistance in data collection.
8. REFERENCES
[1] de Vries, L., Genster, S., & Leeflang, P. S. H. (2012).
Popularity of brand posts on brand fan pages: An
investigation of the effects of social media marketing. Jour.
Interactive Marketing, 26(2), 83-91.
[2] Edelman, D. C. (2010). Branding in the digital age: You're
spending your money in all the wrong places. Harvard
Business Review, Dec. 2010 issue, 62-69.
[3] Freelon, D. (2010). ReCal: Intercoder reliability calculation
as a web service. Int. Jour. Internet Science, 5(1), 20-33.
[4] Goh, K.-Y., Heng, C.-S., Lin, Z. (2013). Social media brand
community and consumer behavior: Quantifying the relative
impact of user- and marketer-generated content. Information
Systems Research, 24(1), 88-107.
[5] Hannaa, R., Rohma, A., & Crittendenb, V. L. (2011). We're
all connected: The power of the social media ecosystem.
Business Horizons, 54(3), 265-273.
[6] Jang, H., Olfman, L., Ko, I., Koh, J., & Kim, K. (2008). The
influence of on-line brand community characteristics on
community commitment and brand loyalty. Int. Jour.
Electronic Commerce, 12(3), 57-80.
[7] Judson, K. M., Devasagayam, P. R., & Buff, C. L. (2012).
Self-perceived brand relevance of and satisfaction with social
media. Marketing Management Journal, 22(2), 131-144.
[8] Laroche, M., Habibi, M. R., Richard, M.-O., &
Sankaranarayanan, R. (2012). The effects of social media
based brand communities on brand community makers, value
creation practices, brand trust and brand loyalty. Computers
in Human Behavior, 28, 1755-1767.
[9] Palmer, A., & Koenig-Lewis, N. (2009). An experiential,
social network-based approach to direct marketing. Direct
Marketing, 3(3), 162-176.
[10] Peruta, A., Ryan, W., & Acquavella, G. (2012). Social media
branding: Comparing brand Facebook pages and website.
Conf. Communication Media, Technology, and Design 2012.
[11] Peruta, A., Ryan, W., & Engelsman, R. (2013). Social media
branding strategies for higher education. Int. Jour.
Technology, Knowledge, and Society, 9(1), 11-23.
[12] Sicilia, M., & Palazón, M. (2008). Brand communities on the
internet. A case study of Coca-Cola's Spanish virtual
community. Corporate Communications, 13(3), 256-270.
[13] Smith, A. N., Fischer, E., & Yongjian, C. (2012). How does
brand-related user-generated content differ across Youtube,
Facebook, and Twitter? Jour. International Marketing, 26,
102-113.